Obese men have comorbid conditions including insulin resistance, dyslipidemia and cardiovascular disease. They also have lower testosterone (T) levels than non-obese men. Administration of T to abdominally obese men with low-normal T levels by one group increased insulin sensitivity and decreased visceral fat. Cardioprotective high-density lipoprotein cholesterol (HDL-C) however, decreased, suggesting that developing selective androgen receptor modulators (SARMs) that are neutral to, or improve the lipid profile are required if T is to be considered a treatment for obesity. This is a prospective, randomized, double-blind trial to determine the effects of T and its metabolites, estradiol (E2) and 5alpha-dihydrotestosterone (DHT) on insulin sensitivity, body composition and lipid metabolism in obese men with low or low-normal T levels. Men between the ages of 21 and 50 years with a body mass index greater than 30 kg/m2, waist measurement greater than 100cm and a waist-to-hip ratio greater than 1.0 will have T clamped within the normal range using a gonadotropin releasing-hormone (GnRH) antagonist and exogenous T except in the placebo group. Sixty subjects will be randomized into one of 4 groups: 1) Placebo group; 2) T alone group- GnRH antagonist + exogenous T + oral placebo; 3) Aromatase inhibitor group- GnRH antagonist + T + anastrazole; and 4) 5alpha-reductase inhibitor group - GnRH antagonist + T + dutasteride. Group 3 will examine outcomes in the absence of E2 while group 4 will examine outcomes in the absence of DHT. Insulin sensitivity by the frequently-sampled intravenous glucose tolerance test, total fat by dual-energy X-ray absorptiometry, regional fat including subcutaneous, visceral and intermuscular fat by magnetic resonance imaging, and intramyocellular fat by 1H-nuclear magnetic resonance spectroscopy will be examined before and after treatment. Fat and muscle biopsies will be obtained before and after treatment to examine tissue-specific insulin sensitivity and to begin an analysis of altered gene expression as it pertains to insulin sensitivity. Lipid metabolism will be examined in detail before and after treatment to include HDL-C, lowdensity lipoprotein size and activities of hepatic lipase, lipoprotein lipase, cholesteryl ester transfer protein and phospholipid transfer protein. These outcomes should begin to describe steroid hormone effects on important metabolic outcomes important for the development of SARMs.